In the aircraft manufacturing process, the tolerance allocation of complex components has an extremely important impact on the life cycle of the aircraft. With the increasing demand for aircraft assembly performance and reliability, tolerance allocation has become one of the most concerned issues for process designers. However, the traditional cost-tolerance models hardly took into account the effects of multiple alternative manufacturing processes. The global search strategy and convergence of optimization methods of the model were insufficient, and it was difficult to obtain more optimized results. Therefore, this paper proposed a novel cost-tolerance model that considers the impact of multiple alternative manufacturing processes on component manufacturing cost and quality loss. Then, a hybrid optimization algorithm combining Monte Carlo simulation and self-adaptive differential evolution (SADE) was presented to achieve cost minimization while ensuring high assembly accuracy. The Monte Carlo simulation was used to solve the problem of tolerance superposition and provided a proper initial population for SADE, which can accelerate the convergence speed and enhances the robustness of the SADE. Moreover, based on the traditional SADE, a new mutation strategy, which expands the diversity of the population, was proposed in combination with the Levy distribution. Improved SADE algorithm had a good optimization effect for large space, nonlinear and non-derivable discrete problems. Finally, a case study of aircraft door assembly was illustrated to verify the effectiveness of the proposed model. The experimental results show the advantage of the method in achieving the cost reduction compared with traditional method.